<p>Deception detection is generally difficult, with people performing only slightly above chance. However, detection accuracy may improve in ecologically valid contexts where judges have professional expertise and contextual familiarity. We examined whether emergency response center (ERC) dispatchers can distinguish between deceptive and truthful emergency calls and which cues guide their judgments. In a two-part perception experiment, 31 Finnish ERC dispatchers judged the veracity of authentic emergency calls comprising hoax (deceptive) and genuine calls (truthful) and reported which cues influenced their decisions. They also rated the credibility of prosodically manipulated call samples where pitch and speech rate were increased and decreased. We applied Signal Detection Theory within generalized linear mixed-effects models to estimate discrimination ability and response bias while accounting for variability across judges and stimuli. Dispatchers achieved 66% accuracy and performed better than typically observed in laboratory studies. However, they showed a truth bias and were more accurate at identifying genuine calls (81%) than hoax calls (53%). Although dispatchers reported relying on content-related cues, self-reported reliance on nervousness, stress, and speech clarity was most consistently associated with veracity judgments. However, none of them predicted accuracy. Lowering pitch and speech rate reduced perceived credibility, but these manipulations affected genuine and hoax calls similarly and did not improve discrimination. Professional expertise and contextual familiarity provided certain advantages, but did not eliminate truth bias or self-reported reliance on cues that did not predict accuracy. These findings suggest that credibility judgments in operational contexts have capabilities and limitations relevant to dispatcher training and practice.</p>